Thyroid carcinoma is one of the fastest growing cancers in the United States with an estimated 44,670 new cases in 2010, of which 80-90% are papillary in type. With current multimodality therapies, 5-year survival is >95%, however prognosis is markedly influenced by patient age and older age at diagnosis is one of the strongest risk factor for differentiated thyroid cancer (DTC) tumor-related death. Age at presentation is such a determinant of prognosis that it is built into the current AJCC TNM thyroid cancer staging scheme with age >45 years set as the threshold for upstaging patients who have similar tumor size and nodal status. Indeed, 10- year DTC survival is 99.8% in patients with stage I disease and only 40.7% in those with stage IV disease. Despite the age-related differences in DTC prognosis, current treatment is characterized by protocols that are rarely modified for elderly patients. Recent knowledge of the genetic changes in thyroid carcinogenesis has led to a better understanding of DTC clinical behavior. An activating mutation of BRAF is now recognized to be frequently associated with aggressive biologic features including a higher risk of disease recurrence and mortality. But current clinical treatment does not yet take BRAF status into account, nor is testing for BRAF widely available. At our highvolume center (~500 cases of thyroid cancer/year) we have routinely tested cytology specimens and thyroid tissue for BRAFV600E since 2007. In this proposal, we hypothesize that older age is a risk modifier for BRAFpositive thyroid cancer. We aim to 1) characterize the association between age, BRAF status and disease outcomes in a large cohort study with longitudinal follow-up, 2) identify novel clinical and molecular markers that are BRAF-dependent and independent risk modifiers and also contribute to the poor disease outcome observed in elderly patients and 3) prospectively study how augmented risk stratification using age and BRAF status can alter current surgical treatment algorithms and improve disease-specific outcomes.